Good day

 

I have a Kodak OG5.48 off-grid inverter. I have been running it with a Hubble AM2 (5kW) lithium battery, as a power backup system. I finally had 4 x 500W Art Solar panels connected last week and have been trying to run with A Solar / Battery / Utility priority order.

I noted that the inverter does not regularly accept the solar power coming in and the MPPT controller icon just flashes on the LCD display. I cannot find this scenario in the manual and therefore am not sure how to troubleshoot.

What is the VOC of the panels? How are they connected? Series or parallel? 

 

What is working range of the MPPT? And does your voltage reading fall within this parameter?

 

The inverter will only prioritize PV if it deemed stable and withing the working range. Ive seen my inverter take up to 5 minutes to switch to PV after its been to Utility

Thank you kindly for your reply. The input voltage range is 120V - 450V with a max open circuit of 500V.

I have 4 panels connected in series. The nominal voltage per panel is 35.93V.

I have a circuit breaker inline between the panels and the inverter (about 1m away from the inverter, so it should cover most of the voltage drop due to cable length) where I measured about 65V yesterday, around the middle of the day. So I suspect that it is a voltage problem and I will need to add more panels to increase the voltage to the inverter. Unless there is some other unforeseen element that I'm missing.

Yes, your PV voltage is too low so the MPPT doesn't start working. You need at least 120V for it to kick in.

46 minutes ago, DinoV said:

Thank you kindly for your reply. The input voltage range is 120V - 450V with a max open circuit of 500V.

I have 4 panels connected in series. The nominal voltage per panel is 35.93V.

I have a circuit breaker inline between the panels and the inverter (about 1m away from the inverter, so it should cover most of the voltage drop due to cable length) where I measured about 65V yesterday, around the middle of the day. So I suspect that it is a voltage problem and I will need to add more panels to increase the voltage to the inverter. Unless there is some other unforeseen element that I'm missing.

@DinoV65v doesn't make sense? Disconnect the panels from the Mppt and measure the open curcuit voltage should be close to 143v measure at 10 to 11 am.

Thanks everyone.

When I have a free moment, I'll climb up to the panels themselves and measure the VOC directly. The VOC of 65V was double checked this morning (between 10-11am, at the inverter) and confirmed. I don't have ideal orientation on my panels (closer to NE) and there is about 8m - 10m of cabling from the inverter to the panel array. So I suspect I'm experiencing more voltage loss due to those.

Out of interest, let's say that 65V (55% loss) is ridiculous. What sort of inefficiency should I be expecting from poor orientation and voltage loss over distance?

Before you climb up to measure, make sure you switched off the PV (isolator and/or fuses). You don't want to mess with high voltage DC. Do you have any shading on the panels during that time period? Also, do you know your PV cable thickness? 4mm? 6mm?

For sure - isolator in place.

Very little shading until after 2pm (Winter months)

4mm cable currently used

11 minutes ago, DinoV said:

Thanks everyone.

When I have a free moment, I'll climb up to the panels themselves and measure the VOC directly. The VOC of 65V was double checked this morning (between 10-11am, at the inverter) and confirmed. I don't have ideal orientation on my panels (closer to NE) and there is about 8m - 10m of cabling from the inverter to the panel array. So I suspect I'm experiencing more voltage loss due to those.

Out of interest, let's say that 65V (55% loss) is ridiculous. What sort of inefficiency should I be expecting from poor orientation and voltage loss over distance?

you wont have so much loss due to cable lengths. 

i personally think the panels are connected 2P2S perhaps. 

even at a very bad angle, the VOC should not be much lower. as soon as you start producing the volts might drop very quickly - no stamina. 

3 hours ago, DinoV said:

I have 4 panels connected in series. The nominal voltage per panel is 35.93V.

Assuming you have ARTsolar – 500 Watt Solar Panel – Mono Percium Full Black then you should be getting around 180Voc mid morning. 65Voc is just too low.

I have to agree with @Flouw that you probably have 2s2p sting configuration instead of 4s1p.

Thanks good to know.

Alas, the panels are all in series - positive of one to the negative of the next - all the way through.

EDIT

I'll make time to take more readings on my multimeter before purchasing the two extra panels. I'm certainly not keen on spending more at this stage plus the next sub-structure installation will be more difficult than the current one, so it will be a bit of a headache to add more panels.

Edited July 18, 20232 yr by DinoV

is there ANY shading on the panels at all? 

having some shade on any of the panels in series will have a major voltage drop.

did you measure VOC with the circuit breaker disconnected. 

 

last but not least, have you seen any PV charging happening since the install? and if you put the inverter into battery only mode. 

when its not prioritized, they tend to be finicky as to when to use PV.

I have a clear sky scenario here until 2pm during Winter - not factoring in clouds of course. However since starting this post, there hasn't been a cloud in sight for 2-3 days. (clear, clear, clear).

I have received some input from the panels since connecting last week. It just seems very limited. I just checked the display and it said 5.5kWh. Not sure if that's today only OR since connecting (i.e. total historical production).

Also, while cycling through the display, I note that the inverter also offers a voltage reading for the PV input - currently 71V at 1:46pm

that voltage still seems very low for 4 panels in series. 

 

dont you have any monitoring app connected to see whats happening?

perhaps a dud panel? 

 

i would start by disconnecting and testing each panel individually. something is off.

So I measured each panel this morning at 10:30am. Each one seems fine with the results ranging between 42V and 44V (It's partly cloudy today, mostly cloudy).

Back at the breaker (8m cable length to the neutral end of the PV series and 4m to the Live end of the series) the voltage measured 80V. The only thing between the DC circuit breaker and the panels is cabling.

If I understand this correctly, I should be getting ±160V across the panels but am only receiving 80V after the cable run to the breaker. That's a 50% drop in voltage over the distance. I'm not an electrician but this does seem extreme with my limited understanding.

The other thing that confuses me a little is that it is cloudy today and I appear to have a higher voltage reading compared to the 65V I had yesterday which was perfectly cloudless.

Do you have any way to measure the combined string on the roof before it disappears into the cable and head towards the circuit breaker? That way you'll know if it's the cabling or the panels.

Edit: in fact, do that for 2, 3, 4 panels in series and check where the voltage drops.

Edited July 19, 20232 yr by p_i

I was also wondering about that. The problem is that the "+" of the first panel is 4m away from the "-" of the last panel. My little multi-meter can't stretch that far.

You should be able to at least measure 2 panels in series at a time. Maybe that'll give you an indication of whether the problem lies with the panels or elsewhere.

1 hour ago, DinoV said:

That's a 50% drop in voltage over the distance. I'm not an electrician but this does seem extreme with my limited understanding

Volt drop calculation for 4mm² copper cable running dc through it. Single run of 20 meters x 2 = 40meters. Cable temprature 50 deg C ,Dc voltdrop = 2.67 volts DC voltdrop 1.48%. You have a problem with either one of your panels or high resistance on cabling or fuse disconnect.

Edited July 19, 20232 yr by TaliaB

Below link for ac or dc calculator to determine volt drop in specific cable size.

https://photovoltaic-software.com/solar-tools/voltage-drop-calculator-dc-ac

Thanks @TaliaB I'm leaning towards a resistance problem on the cable. I will still try find a way of testing the total string of panels (they seemed fine individually when i measured them earlier). It's a straight run from the panels to my breaker. There is only 1 x male and 1 x female connectors besides the cable - nothing else inline.

I found that voltage drop calculator earlier, after taking the measurements I posted above and noting the massive voltage drop, which also seems to point to the cables being a problem. It is an older cable that was already here previously connecting a smaller 280W panel. I removed the panel but didn't change the cable. So it could be that there is an unseen break in the cable.

I'll tinker more on Saturday and see if I can test resistance across the cable. I'll be happier if it's a cable fault as it will be cheaper than buying extra PV panels.

55 minutes ago, DinoV said:

I'll tinker more on Saturday and see if I can test resistance across the cable.

Disconnect the cables pos and neg at the panels and join the wire on the roof with a connection block. Go down to the dc breaker switch the breaker off and measure on top of the breaker setting your tester on resistance (omh scale x1) put the leads together and check the resistance of the leads then measure the resistance on the wire and subtract the first reading from the resistance measured on the breaker. Then disconnect the wires at the bottom of the breaker then switch the breaker on and measure at the bottom of the breaker the resistance again. Record your reading there should be no diffrence between the 2 readings. Your resistance reading should be close to 0 omhs for your distance as 4mm² copper wire is 4.3 ohms/km.

Edited July 19, 20232 yr by TaliaB

I assume that the ≈80V reading is under load. What is the voltage at the junction box with no load?

My suspicion is that one or more panels or connections is high resistance. If so,  you would get ≈160 V with no load, but as soon as there is load, one or two panels drops to near zero volts.

Adding extra panels in series won't help this situation.

4 hours ago, Coulomb said:

but as soon as there is load, one or two panels drops to near zero volts

@DinoV Another way to test the panels is disconnect all the panels from each other. If your multimeter can measure amps normally max 10 amps dc. You need to change the pos probe to the current input. If your meter has 3 probe inputs it can measure amps. Then on amps put the leads across the pos and neg wire there is going to be a small arc so do it quickly and positively then check the amps on your reading and compare it with your panel spec ISC they sould be very close. Do this in full sunshine at around 11 am. I do this test as default before installing the panels i use 2 multimeters one for current(ISC) and then check the VMP to ensure the panels are close to STC values.

Edited July 20, 20232 yr by TaliaB
Spelling

Thanks @Coulomb. Unless there is some unknown definition of "under load", I'm pretty certain that I'm measuring without it being under load. I measured the panels directly while being disconnected. I think this means that it's not under load. Then at the junction box (approx 8m away from neutral of the PV series), before connecting to the breaker (breaker off, which again I think means no loads) I measured again. So each panel was in the range of 42V - 44V and then open circuit reading at the junction box was 80V (no load). The inverter itself gave a slightly higher reading (89V) than my multimeter when everything was connected, switched on and under load. I do agree, that at present, adding extra panels doesn't seem to be the solution. If the panels themselves are high resistance what would be the solution?

@TaliaB I was wondering about measuring current. So I will add this to my list of "things to measure" on Saturday. Thank you so much.